RESUMEN
OBJECTIVES: To develop and test zone-specific prostate-specific antigen density (sPSAD) combined with PI-RADS to guide prostate biopsy decision strategies (BDS). METHODS: This retrospective study included consecutive patients, who underwent prostate MRI and biopsy (01/2012-10/2018). The whole gland and transition zone (TZ) were segmented at MRI using a retrained deep learning system (DLS; nnU-Net) to calculate PSAD and sPSAD, respectively. Additionally, sPSAD and PI-RADS were combined in a BDS, and diagnostic performances to detect Grade Group ≥ 2 (GG ≥ 2) prostate cancer were compared. Patient-based cancer detection using sPSAD was assessed by bootstrapping with 1000 repetitions and reported as area under the curve (AUC). Clinical utility of the BDS was tested in the hold-out test set using decision curve analysis. Statistics included nonparametric DeLong test for AUCs and Fisher-Yates test for remaining performance metrics. RESULTS: A total of 1604 patients aged 67 (interquartile range, 61-73) with 48% GG ≥ 2 prevalence (774/1604) were evaluated. By employing DLS-based prostate and TZ volumes (DICE coefficients of 0.89 (95% confidence interval, 0.80-0.97) and 0.84 (0.70-0.99)), GG ≥ 2 detection using PSAD was inferior to sPSAD (AUC, 0.71 (0.68-0.74)/0.73 (0.70-0.76); p < 0.001). Combining PI-RADS with sPSAD, GG ≥ 2 detection specificity doubled from 18% (10-20%) to 43% (30-44%; p < 0.001) with similar sensitivity (93% (89-96%)/97% (94-99%); p = 0.052), when biopsies were taken in PI-RADS 4-5 and 3 only if sPSAD was ≥ 0.42 ng/mL/cc as compared to all PI-RADS 3-5 cases. Additionally, using the sPSAD-based BDS, false positives were reduced by 25% (123 (104-142)/165 (146-185); p < 0.001). CONCLUSION: Using sPSAD to guide biopsy decisions in PI-RADS 3 lesions can reduce false positives at MRI while maintaining high sensitivity for GG ≥ 2 cancers. CLINICAL RELEVANCE STATEMENT: Transition zone-specific prostate-specific antigen density can improve the accuracy of prostate cancer detection compared to MRI assessments alone, by lowering false-positive cases without significantly missing men with ISUP GG ≥ 2 cancers. KEY POINTS: ⢠Prostate biopsy decision strategies using PI-RADS at MRI are limited by a substantial proportion of false positives, not yielding grade group ≥ 2 prostate cancer. ⢠PI-RADS combined with transition zone (TZ)-specific prostate-specific antigen density (PSAD) decreased the number of unproductive biopsies by 25% compared to PI-RADS only. ⢠TZ-specific PSAD also improved the specificity of MRI-directed biopsies by 9% compared to the whole gland PSAD, while showing identical sensitivity.
RESUMEN
Background Clinically significant prostate cancer (PCa) diagnosis at MRI requires accurate and efficient radiologic interpretation. Although artificial intelligence may assist in this task, lack of transparency has limited clinical translation. Purpose To develop an explainable artificial intelligence (XAI) model for clinically significant PCa diagnosis at biparametric MRI using Prostate Imaging Reporting and Data System (PI-RADS) features for classification justification. Materials and Methods This retrospective study included consecutive patients with histopathologic analysis-proven prostatic lesions who underwent biparametric MRI and biopsy between January 2012 and December 2017. After image annotation by two radiologists, a deep learning model was trained to detect the index lesion; classify PCa, clinically significant PCa (Gleason score ≥ 7), and benign lesions (eg, prostatitis); and justify classifications using PI-RADS features. Lesion- and patient-based performance were assessed using fivefold cross validation and areas under the receiver operating characteristic curve. Clinical feasibility was tested in a multireader study and by using the external PROSTATEx data set. Statistical evaluation of the multireader study included Mann-Whitney U and exact Fisher-Yates test. Results Overall, 1224 men (median age, 67 years; IQR, 62-73 years) had 3260 prostatic lesions (372 lesions with Gleason score of 6; 743 lesions with Gleason score of ≥ 7; 2145 benign lesions). XAI reliably detected clinically significant PCa in internal (area under the receiver operating characteristic curve, 0.89) and external test sets (area under the receiver operating characteristic curve, 0.87) with a sensitivity of 93% (95% CI: 87, 98) and an average of one false-positive finding per patient. Accuracy of the visual and textual explanations of XAI classifications was 80% (1080 of 1352), confirmed by experts. XAI-assisted readings improved the confidence (4.1 vs 3.4 on a five-point Likert scale; P = .007) of nonexperts in assessing PI-RADS 3 lesions, reducing reading time by 58 seconds (P = .009). Conclusion The explainable AI model reliably detected and classified clinically significant prostate cancer and improved the confidence and reading time of nonexperts while providing visual and textual explanations using well-established imaging features. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Chapiro in this issue.
Asunto(s)
Aprendizaje Profundo , Neoplasias de la Próstata , Masculino , Humanos , Anciano , Próstata/patología , Neoplasias de la Próstata/patología , Imagen por Resonancia Magnética/métodos , Inteligencia Artificial , Estudios RetrospectivosRESUMEN
OBJECTIVES: The goal of this study was to investigate the effects of TACE using Lipiodol, Oncozene™ drug-eluting embolics (DEEs), or LUMI™-DEEs alone, or combined with bicarbonate on the metabolic and immunological tumor microenvironment in a rabbit VX2 tumor model. METHODS: VX2 liver tumor-bearing rabbits were assigned to five groups. MRI and extracellular pH (pHe) mapping using Biosensor Imaging of Redundant Deviation in Shifts (BIRDS) were performed before and after intra-arterial therapy with conventional TACE (cTACE), DEE-TACE with Idarubicin-eluting Oncozene™-DEEs, or Doxorubicin-eluting LUMI™-DEEs, each with or without prior bicarbonate infusion, and in untreated rabbits or treated with intra-arterial bicarbonate only. Imaging results were validated with immunohistochemistry (IHC) staining of cell viability (PCNA, TUNEL) and immune response (HLA-DR, CD3). Statistical analysis was performed using Mann-Whitney U test. RESULTS: pHe mapping revealed that combining cTACE with prior bicarbonate infusion significantly increased tumor pHe compared to control (p = 0.0175) and cTACE alone (p = 0.0025). IHC staining revealed peritumoral accumulation of HLA-DR+ antigen-presenting cells and CD3 + T-lymphocytes in controls. cTACE-treated tumors showed reduced immune infiltration, which was restored through combination with bicarbonate. DEE-TACE with Oncozene™-DEEs induced moderate intratumoral and marked peritumoral infiltration, which was slightly reduced with bicarbonate. Addition of bicarbonate prior to LUMI™-beads enhanced peritumoral immune cell infiltration compared to LUMI™-beads alone and resulted in the strongest intratumoral immune cell infiltration across all treated groups. CONCLUSIONS: The choice of chemoembolic regimen for TACE strongly affects post-treatment TME pHe and the ability of immune cells to accumulate and infiltrate the tumor tissue. KEY POINTS: ⢠Combining conventional transarterial chemotherapy with prior bicarbonate infusion increases the pHe towards a more physiological value (p = 0.0025). ⢠Peritumoral infiltration and intratumoral accumulation patterns of antigen-presenting cells and T-lymphocytes after transarterial chemotherapy were dependent on the choice of the chemoembolic regimen. ⢠Combination of intra-arterial treatment with Doxorubicin-eluting LUMI™-beads and bicarbonate infusion resulted in the strongest intratumoral presence of immune cells (positivity index of 0.47 for HLADR+-cells and 0.62 for CD3+-cells).
Asunto(s)
Carcinoma Hepatocelular , Quimioembolización Terapéutica , Neoplasias Hepáticas , Animales , Carcinoma Hepatocelular/patología , Quimioembolización Terapéutica/métodos , Doxorrubicina , Aceite Etiodizado , Neoplasias Hepáticas/patología , Conejos , Microambiente TumoralRESUMEN
OBJECTIVES: To train a deep learning model to differentiate between pathologically proven hepatocellular carcinoma (HCC) and non-HCC lesions including lesions with atypical imaging features on MRI. METHODS: This IRB-approved retrospective study included 118 patients with 150 lesions (93 (62%) HCC and 57 (38%) non-HCC) pathologically confirmed through biopsies (n = 72), resections (n = 29), liver transplants (n = 46), and autopsies (n = 3). Forty-seven percent of HCC lesions showed atypical imaging features (not meeting Liver Imaging Reporting and Data System [LI-RADS] criteria for definitive HCC/LR5). A 3D convolutional neural network (CNN) was trained on 140 lesions and tested for its ability to classify the 10 remaining lesions (5 HCC/5 non-HCC). Performance of the model was averaged over 150 runs with random sub-sampling to provide class-balanced test sets. A lesion grading system was developed to demonstrate the similarity between atypical HCC and non-HCC lesions prone to misclassification by the CNN. RESULTS: The CNN demonstrated an overall accuracy of 87.3%. Sensitivities/specificities for HCC and non-HCC lesions were 92.7%/82.0% and 82.0%/92.7%, respectively. The area under the receiver operating curve was 0.912. CNN's performance was correlated with the lesion grading system, becoming less accurate the more atypical imaging features the lesions showed. CONCLUSION: This study provides proof-of-concept for CNN-based classification of both typical- and atypical-appearing HCC lesions on multi-phasic MRI, utilizing pathologically confirmed lesions as "ground truth." KEY POINTS: ⢠A CNN trained on atypical appearing pathologically proven HCC lesions not meeting LI-RADS criteria for definitive HCC (LR5) can correctly differentiate HCC lesions from other liver malignancies, potentially expanding the role of image-based diagnosis in primary liver cancer with atypical features. ⢠The trained CNN demonstrated an overall accuracy of 87.3% and a computational time of < 3 ms which paves the way for clinical application as a decision support instrument.
Asunto(s)
Carcinoma Hepatocelular , Aprendizaje Profundo , Neoplasias Hepáticas , Carcinoma Hepatocelular/diagnóstico por imagen , Medios de Contraste , Humanos , Neoplasias Hepáticas/diagnóstico por imagen , Imagen por Resonancia Magnética , Estudios RetrospectivosRESUMEN
OBJECTIVES: To develop and validate a proof-of-concept convolutional neural network (CNN)-based deep learning system (DLS) that classifies common hepatic lesions on multi-phasic MRI. METHODS: A custom CNN was engineered by iteratively optimizing the network architecture and training cases, finally consisting of three convolutional layers with associated rectified linear units, two maximum pooling layers, and two fully connected layers. Four hundred ninety-four hepatic lesions with typical imaging features from six categories were utilized, divided into training (n = 434) and test (n = 60) sets. Established augmentation techniques were used to generate 43,400 training samples. An Adam optimizer was used for training. Monte Carlo cross-validation was performed. After model engineering was finalized, classification accuracy for the final CNN was compared with two board-certified radiologists on an identical unseen test set. RESULTS: The DLS demonstrated a 92% accuracy, a 92% sensitivity (Sn), and a 98% specificity (Sp). Test set performance in a single run of random unseen cases showed an average 90% Sn and 98% Sp. The average Sn/Sp on these same cases for radiologists was 82.5%/96.5%. Results showed a 90% Sn for classifying hepatocellular carcinoma (HCC) compared to 60%/70% for radiologists. For HCC classification, the true positive and false positive rates were 93.5% and 1.6%, respectively, with a receiver operating characteristic area under the curve of 0.992. Computation time per lesion was 5.6 ms. CONCLUSION: This preliminary deep learning study demonstrated feasibility for classifying lesions with typical imaging features from six common hepatic lesion types, motivating future studies with larger multi-institutional datasets and more complex imaging appearances. KEY POINTS: ⢠Deep learning demonstrates high performance in the classification of liver lesions on volumetric multi-phasic MRI, showing potential as an eventual decision-support tool for radiologists. ⢠Demonstrating a classification runtime of a few milliseconds per lesion, a deep learning system could be incorporated into the clinical workflow in a time-efficient manner.
Asunto(s)
Carcinoma Hepatocelular/diagnóstico por imagen , Aprendizaje Profundo , Neoplasias Hepáticas/diagnóstico por imagen , Redes Neurales de la Computación , Adulto , Anciano , Neoplasias de los Conductos Biliares/diagnóstico por imagen , Conductos Biliares Intrahepáticos , Colangiocarcinoma/diagnóstico por imagen , Femenino , Humanos , Interpretación de Imagen Asistida por Computador/métodos , Imagen por Resonancia Magnética/métodos , Masculino , Persona de Mediana Edad , Curva ROC , Reproducibilidad de los Resultados , Sensibilidad y Especificidad , Estados UnidosRESUMEN
OBJECTIVES: To develop a proof-of-concept "interpretable" deep learning prototype that justifies aspects of its predictions from a pre-trained hepatic lesion classifier. METHODS: A convolutional neural network (CNN) was engineered and trained to classify six hepatic tumor entities using 494 lesions on multi-phasic MRI, described in Part 1. A subset of each lesion class was labeled with up to four key imaging features per lesion. A post hoc algorithm inferred the presence of these features in a test set of 60 lesions by analyzing activation patterns of the pre-trained CNN model. Feature maps were generated that highlight regions in the original image that correspond to particular features. Additionally, relevance scores were assigned to each identified feature, denoting the relative contribution of a feature to the predicted lesion classification. RESULTS: The interpretable deep learning system achieved 76.5% positive predictive value and 82.9% sensitivity in identifying the correct radiological features present in each test lesion. The model misclassified 12% of lesions. Incorrect features were found more often in misclassified lesions than correctly identified lesions (60.4% vs. 85.6%). Feature maps were consistent with original image voxels contributing to each imaging feature. Feature relevance scores tended to reflect the most prominent imaging criteria for each class. CONCLUSIONS: This interpretable deep learning system demonstrates proof of principle for illuminating portions of a pre-trained deep neural network's decision-making, by analyzing inner layers and automatically describing features contributing to predictions. KEY POINTS: ⢠An interpretable deep learning system prototype can explain aspects of its decision-making by identifying relevant imaging features and showing where these features are found on an image, facilitating clinical translation. ⢠By providing feedback on the importance of various radiological features in performing differential diagnosis, interpretable deep learning systems have the potential to interface with standardized reporting systems such as LI-RADS, validating ancillary features and improving clinical practicality. ⢠An interpretable deep learning system could potentially add quantitative data to radiologic reports and serve radiologists with evidence-based decision support.
Asunto(s)
Carcinoma Hepatocelular/diagnóstico por imagen , Aprendizaje Profundo , Neoplasias Hepáticas/diagnóstico por imagen , Redes Neurales de la Computación , Adulto , Anciano , Algoritmos , Neoplasias de los Conductos Biliares/diagnóstico por imagen , Conductos Biliares Intrahepáticos , Colangiocarcinoma/diagnóstico por imagen , Femenino , Humanos , Interpretación de Imagen Asistida por Computador/métodos , Aprendizaje Automático , Imagen por Resonancia Magnética/métodos , Masculino , Persona de Mediana Edad , Valor Predictivo de las Pruebas , Prueba de Estudio Conceptual , Estudios RetrospectivosRESUMEN
PURPOSE: The ever-increasing volume of medical imaging data and interest in Big Data research brings challenges to data organization, categorization, and retrieval. Although the radiological value chain is almost entirely digital, data structuring has been widely performed pragmatically, but with insufficient naming and metadata standards for the stringent needs of image analysis. To enable automated data management independent of naming and metadata, this study focused on developing a convolutional neural network (CNN) that classifies medical images based solely on voxel data. METHOD: A 3D CNN (3D-ResNet18) was trained using a dataset of 31,602 prostate MRI volumes with 10 different sequence types of 1243 patients. A five-fold cross-validation approach with patient-based splits was chosen for training and testing. Training was repeated with a gradual reduction in training data assessing classification accuracies to determine the minimum training data required for sufficient performance. The trained model and developed method were tested on three external datasets. RESULTS: The model achieved an overall accuracy of 99.88 % ± 0.13 % in classifying typical prostate MRI sequence types. When being trained with approximately 10 % of the original cohort (112 patients), the CNN still achieved an accuracy of 97.43 % ± 2.10 %. In external testing the model achieved sensitivities of > 90 % for 10/15 tested sequence types. CONCLUSIONS: The herein developed CNN enabled automatic and reliable sequence identification in prostate MRI. Ultimately, such CNN models for voxel-based sequence identification could substantially enhance the management of medical imaging data, improve workflow efficiency and data quality, and allow for robust clinical AI workflows.
Asunto(s)
Metadatos , Próstata , Masculino , Humanos , Imagen por Resonancia Magnética , Redes Neurales de la Computación , Procesamiento de Imagen Asistido por Computador/métodosRESUMEN
Dual-energy computed tomography (DECT) is an imaging technique that combines nondestructive morphological cross-sectional imaging of objects and the quantification of their chemical composition. However, its potential to assist investigations in paleontology has not yet been explored. This study investigates quantitative DECT for the nondestructive density- and element-based material decomposition of fossilized bones. Specifically, DECT was developed and validated for imaging-based calcium and fluorine quantification in bones of five fossil vertebrates from different geological time periods and of one extant vertebrate. The analysis shows that DECT material maps can differentiate bone from surrounding sediment and reveals fluorine as an imaging marker for fossilized bone and a reliable indicator of the age of terrestrial fossils. Moreover, the jaw bone mass of Tyrannosaurus rex showed areas of particularly high fluorine concentrations on DECT, while conventional CT imaging features supported the diagnosis of chronic osteomyelitis. These findings highlight the relevance of radiological imaging techniques in the natural sciences by introducing quantitative DECT imaging as a nondestructive approach for material decomposition in fossilized objects, thereby potentially adding to the toolbox of paleontological studies.
Asunto(s)
Paleontología , Tomografía Computarizada por Rayos X , Animales , Calcio , Flúor , Tomografía Computarizada por Rayos X/métodos , VertebradosRESUMEN
Our purpose was to identify baseline imaging features in patients with liver cancer that correlate with 90Y distribution on postprocedural SPECT and predict tumor response to transarterial radioembolization (TARE). Methods: This retrospective study was approved by the institutional review board and included 38 patients with hepatocellular carcinoma (HCC) (n = 23; 18/23 men; mean age, 62.39 ± 8.62 y; 34 dominant tumors) and non-HCC hepatic malignancies (n = 15; 9/15 men; mean age, 61.13 ± 11.51 y; 24 dominant tumors) who underwent 40 resin-based TARE treatments (August 2012 to January 2018). Multiphasic contrast-enhanced MRI or CT was obtained before and Bremsstrahlung SPECT within 2 h after TARE. Total tumor volume (cm3) and enhancing tumor volume (ETV [cm3] and % of total tumor volume), and total and enhancing tumor burden (%), were volumetrically assessed on baseline imaging. Up to 2 dominant tumors per treated lobe were analyzed. After multimodal image registration of baseline imaging and SPECT/CT, 90Y distribution was quantified on SPECT as tumor-to-normal-liver ratio (TNR). Response was assessed according to RECIST1.1 and quantitative European Association for the Study of the Liver criteria. Clinical parameters were also assessed. Statistical tests included Mann-Whitney U, Pearson correlation, and linear regression. Results: In HCC patients, high baseline ETV% significantly correlated with high TNR on SPECT, demonstrating greater 90Y uptake in the tumor relative to the liver parenchyma (P < 0.001). In non-HCC patients, a correlation between ETV% and TNR was observed as well (P = 0.039). Follow-up imaging for response assessments within 1-4 mo after TARE was available for 23 patients with 25 treatments. The change of ETV% significantly correlated with TNR in HCC (P = 0.039) but not in non-HCC patients (P = 0.886). Additionally, Child-Pugh class B patients demonstrated significantly more 90Y deposition in nontumorous liver than Child-Pugh A patients (P = 0.021). Conclusion: This study identified ETV% as a quantifiable imaging biomarker on preprocedural MRI and CT to predict 90Y distribution on postprocedural SPECT in HCC and non-HCC. However, the relationship between the preferential uptake of 90Y to the tumor and tumor response after radioembolization could be validated only for HCC.